Check that all models were downloaded

Check that all the directories for the .nc files got made

source_dl <- dir(here("data_raw", "CMIP6")) source_id <- idx$source_id %>% unique() %>% str_to_lower() %>% str_replace_all("-", "_") stop_if_not(!any(!source_id %in% source_dl))

Check that all the corresponding .csv files exist

csvs <- list.files(here('data')) stop_if_not(!any(!paste0(source_id, "_data.csv") %in% csvs))

Ensure that all models are complete

For this analysis, I only want to use models with pr, tas, hfss, and hfls variables in all 5 scenarios (historical, ssp126, ssp245, ssp370, and ssp585)

Calculations

Perform necessary calculations to compare PET and SPEI among models and between models and observed.

For PET, I’m using the “energy-only” method proposed by Milly and Dune (2016) eq. 8:

\[ PET = 0.8(R_n - G) \]

Except that in their notes, they estimate \(R_n -G\) as hfls + hfss after converting to units of mm/day using the latent heat of vaporazation of water, given by their eq. 2:

\[ L_v(T) = 2.501 - 0.002361T \] in MJ/kg

For the observed data and the CMIP6 data from the same period, I calculate 3-month SPEI using precipitation and PET.

I calculate drought duration as number of consecutive months with SPEI ≤ -1. A single drought, therefore, is defined here as a span of consecutive months all with SPEI ≤ -1.

Comparison to Observed Historical

I’ve ranked the CMIP6 models based on the correlations of monthly precipitation and temperature and the p-value from a non-parametric t-test (Wilcoxon rank sum test) for the mean duration of droughts defined as the nubmer of consecutive months with SPEI ≤ -1.

Comparison of observed data to CMIP6 'historical' output
Data only from 1980 to 2015 to match observed.
Source Monthly means1 Drought duration and frequency
precipitation temperature pr tas tasmin tasmax duration (mon) mean ± SD2 number
observed3 1.00 1.00 1.00 1.00 3.4±2.8 24
awi_cm_1_1_mr 0.80 0.50 0.65 0.53 2.5±1.3 30
taiesm1 0.71 0.51 NA NA 2.9±2.5 29
fgoals_g3 0.92 0.56 0.48 0.76 2.2±1.7 36
fgoals_f3_l 0.98 0.71 NA NA 2±1.4 41
cas_esm2_0 0.85 0.61 0.83 0.71 2±1.6 40
cmcc_esm2 0.73 0.46 0.72 0.50 2.2±1.5 34
canesm5 0.42 0.10 0.50 0.17 3.1±2.5 21
cmcc_cm2_sr5 0.55 0.36 NA NA 2.5±2.2 37
access_esm1_5 0.69 0.43 0.63 0.50 2.1±1.9 40
bcc_csm2_mr 0.13 0.38 0.62 0.43 2.2±1.3 37
ec_earth3_veg_lr 0.27 0.04 0.45 0.11 2.6±2.1 38
iitm_esm 0.17 −0.27 NA NA 2.2±1.3 35
cams_csm1_0 0.36 −0.27 NA NA 2±1.2 37
access_cm2 0.02 −0.16 0.34 0.03 2±1.2 43

1 Spearman's rho. Rho < 0.45 highlighted in red.

2 Red indicates signifcant Wilcoxon rank sum test (p < 0.05).

3 Observed data from Xavier et al. (2016)

CMIP model details

Below are validation reports and plots of all data downloaded from each CMIP6 source.

access_cm2

access_esm1_5

awi_cm_1_1_mr

bcc_csm2_mr

cams_csm1_0

canesm5

cas_esm2_0

cmcc_cm2_sr5

cmcc_esm2

ec_earth3_veg_lr

fgoals_f3_l

fgoals_g3

iitm_esm

taiesm1